Sample Size Determination for Bibliographic Retrieval Studies
Author Information
Author(s): Yao Xiaomei, Wilczynski Nancy L, Walter Stephen D, Haynes R Brian
Primary Institution: McMaster University
Hypothesis
What is the minimal number of high-quality articles needed to update or create new MEDLINE search strategies?
Conclusion
Randomly sampling a small subset of journals with sufficient high-quality articles is an efficient way to update or create search strategies for high-quality articles on therapy in MEDLINE.
Supporting Evidence
- The study found that a subset of 15 randomly selected journals was adequate for updating search strategies.
- New search strategies derived from random sampling performed better than those from top journals in low-yielding journal subsets.
- The concentration of high-quality articles was significantly higher in treatment studies compared to diagnosis and prognosis.
Takeaway
To find good medical articles quickly, we can look at a small number of journals instead of searching through many. This helps save time and effort.
Methodology
The study calculated the number of high-quality articles needed based on desired confidence intervals and tested new search strategies in journal subsets.
Potential Biases
There is a risk that randomly selected journals may not always contain enough high-quality articles.
Limitations
The performance of search strategies may vary with different journal subsets, and low concentrations of diagnosis and prognosis articles limit the effectiveness of the approach.
Participant Demographics
The study involved 49,028 articles from 161 journals published in 2000.
Statistical Information
P-Value
p<0.05
Confidence Interval
95%
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website